Variation Characteristics And Prediction Of Runoff And Sediment In The Upper Reaches Of Yarkant River Under Changing Environment

Located in the alpine mountainous area,the upper reaches of the Yarkant River is a typical ice and snow melt supply river.With global warming,the glaciers in the mountainous areas of the basin are thinning and retreating,and the melting of ice and snow is accelerating,resulting in a substantial increase in the uncertainty of water resources.Flood prevention and disaster mitigation work in the river basin has brought great challenges.Studying the variation characteristics and prediction of the water and sediment of the Yarkant River under the changing environment is of great significance for predicting the amount of water and sediment entering the Altash Reservoir in the future and promoting the sustainable use of water resources in the basin.In this paper,the upper reaches of the Yarkant River Basin is the research area.Based on the measured runoff and sediment data of the Kuruklangan Hydrological Station and the Kaqun Hydrological Station,the characteristic analysis of the upstream runoff and sediment series of the basin is carried out by mathematical statistics.Copula theory built a two-dimensional joint distribution model of runoff and sediment,analyzes the changes in the frequency of runoff and sediment in the basin under changing environments,and uses the cumulative skew rate of change comparison method to estimate the contribution of climate change and human activities to changes in runoff and sediment.Finally,the CMIP5 data drive BP neural network to predict the future runoff and sediment.The main research conclusions obtained are as follows:(1)During 1954?2015,the trend of runoff and sediment are basically consistent,the runoff and sediment have been increasing on the whole,and only a few years have shown a downward trend.The runoff and sediment have both changed in 1993,the relationship between runoff and sediment also changed in 1993.(2)The synchrony frequency of runoff and sediment is greater than that of runoff and sediment asynchrony during 1954?2015,and the frequency of rich runoff and poor sediment,poor runoff and rich sediment is 0.In 1954?1992,the highest type frequency is normal-normal(38.46%),followed by poor runoff and normal sand(15.38%).In 1993?2015,the type of normal-normal with the highest encounter frequency(43.48%),followed by the type of dry-dry(30.43%).The synchronization frequency of runoff and sediment during 1993?2015 was 23.64% higher than that of 1954?1992.(3)In 1993?2015,the impact of precipitation on runoff changes is 3.60 billion m~3,accounting for54.23%.The impact of human activities on runoff changes is 3.40 billion m~3,accounting for 45.77%.The impact of precipitation on changes in sediment transport is 137.258 billion m~3,accounting for71.76%;the impact of human activities on runoff changes is 54.027 billion m~3,accounting for28.24%.Climate change factors are the leading factors that change runoff and sediment transport.(4)CMIP5 data drives BP neural network model to predict future runoff and sediment,The prediction results show that: the runoff and sediment forecasts increase slowly under the models of CNRM-CM5 and MIROC-ESM in RCP2.6 and RCP4.5 scenarios,the growth is more significant under the RCP8.5 scenario.The runoff and sediment forecasts of the three scenarios in the Nor ESM1-M model all show a steady growth trend,of which the increase is more obvious under the RCP4.5 scenario,followed by the RCP8.5 scenario...